The applicant's overall objective would be to elucidate the mechanisms of ventricular fibrillation (VF) using a combination of high resolution (500-electrode) mapping and computer simulations. The focus would be on global rather than cellular mechanisms. Hypotheses to be tested are: 1. Reentrant rotors are present during Wigger's stage II VF and are responsible for maintaining VF. 2. The generation and maintenance of rotors depend on myocardial fiber orientation and the influence of the Purkinje fiber network. 3. A critical mass is needed to sustain VF because of the limited life span of the reentrant rotors. The role of the Purkinje network would be evaluated by performing experiments with and without the Purkinje network ablated using Lugol's solution. Fiber orientation would be determined by histopathological examination. The computer simulation studies would be based on a three-dimensional model of myocardial activation using the (modified) FitzHugh-Nagumo system of equations. An existing, detailed finite-element model of the architecture of the canine heart would be used to represent the geometry and structure (i.e. fiber orientation). Model parameters would be obtained by fitting simulated activation patterns in two-dimensional and three-dimensional models to measured normal and reentrant patterns mapped in small regions of the canine heart. The model would be extrapolated to simulate global activation in the whole heart and used to study the effect of heart size on VF and to examine the "critical mass" hypothesis in relation to the number of rotors. The model would also be used to study the role of fiber orientation and of the Purkinje network on rotor formation and maintenance.